Development of a Model for Evaluating Risk of Disaster Due to Scouring Around the Pier Based on Machine Learning

Kabeyama, Yoshihiro; Kadono, Takuma; Okazaki, Yuriko; Nakamura, Taro; Okazaki, Shinichiro; Kajitani, Yoshio; Watanabe, Satoshi; Samizo, Masahiko

doi:10.2208/jscejhe.76.2_i_601

Cited by 4 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Samizo [2] proposed a method to evaluate the risk of local scouring around piers based on the data of existing disasters via a multivariate model using several parameters that can be obtained from maintenance. Additionally, the authors proposed a model for evaluating the disaster around a pier due to scouring using several explanatories as properties of structure and river conditions based on the neural network [3]. On the other hand, during heavy rainfall, it is necessary to conduct risk assessments frequently based on the weather condition so that the government or bridge management committee can determine bridge usagerestrictions and secure evacuation routes for local residences.…”

Section: Introductionmentioning

confidence: 99%

Development of a Model for Evaluating the Disaster Risk Around a Pier due to Local Scouring Based on the Observed Weather Information

Kadono

Okazaki

Kajitani

et al. 2021

Advances in Science and Technology

Self Cite

View full text Add to dashboard Cite

Heavy rainfall disasters frequently damage bridge piers due to scouring, which resulted in collapse of bridges in many areas in Japan. In this study, we developed a model for evaluating the tilting risk of bridge pier due to scouring around the pier, which fluctuates depending on rainfall conditions based on machine learning. For evaluating the risk potential of scouring, we developed a model based on past disaster data due to scouring around the pier using a neural network. Furthermore, a sensitivity analysis was conducted using the parameters of explanatory variables of the developed model, river level, and distance from the water edge to the pier. The results showed that the disaster risk around the pier due to scouring increased with the increase in river level and decrease in the distance from the water edge to the pier. Additionally, a river level prediction model was developed using support vector regression with the precipitation time measured 5 – 8 h beforehand and river level measured several hours earlier as an explanatory variable. Furthermore, this study shows that the two developed models can be combined with each other to assess the disaster risk around the jetty due to scouring, which varies with rainfall conditions, based on the observed meteorological information.

show abstract

Section: Introductionmentioning

confidence: 99%